Ask Your Question

Kailegh's profile - activity

2017-07-13 08:07:07 -0500 received badge  Necromancer (source)
2017-07-13 04:55:41 -0500 answered a question biometria facial

You can try OpenCV example of face recognition: OpenCV Face Recognition

It may not be the most accurate method, but since you state it is for the course completion work I do not really think you need much more than that

2017-07-13 04:48:16 -0500 commented answer Shape alignement and differences computation

I have worked with the 3D version a lot, I did not know it also had a 2D version, I will give it a try Regarding points 2 and 3, especially point 3, do you have any idea? thanks a lot!

2017-07-13 02:28:15 -0500 commented question Shape alignement and differences computation

with 2D images, the ones above are a possible example of what I would get. The possibility of working with 3D pointclouds is open, but for now I want to keep it in 2D

2017-07-12 07:08:56 -0500 asked a question Shape alignement and differences computation

Hi! I intend to scan certain shapes, beam kind, and try to align them to a "ground truth" shape. One this is done I want to compute some differences, for example, compute the verticality of the B line compared to the original and measure the area difference in part A.

image description

To sum up and put things in order:

1.- I should align the scanned shape to the original

2.- Compute the "angle deviation" between the B line scanned and the original. I know it is not really a straight line but I think it can be approximated to one.

3.- The head part A, due to the wear has lost its shape and therefore lost part of the area, I should project it over the original shape and compute the area difference. This is the point that I really do not know how to face.

Do you have any ideas about how to approach this problem? I am not asking you to provide the code, just guide me in what things I can try to solve this

Thanks a lot for your help!!!!

2015-07-21 05:07:59 -0500 answered a question techniques available for image recognition in screenshots

look at this post, i may help you: http://answers.opencv.org/question/86...

2015-07-10 11:59:51 -0500 commented question Best method to recognize objects?

in thar module it uses surf detector, i tried that but i havent got good results, i have to calculate descripitors of all traafic roads and compare them my imae detected and the smalles distance is the best result? because I tried that once and it did work quite well

2015-07-10 07:35:52 -0500 commented question Best method to recognize objects?

could you explain me a bit how i do that? or where cain i find a tutorial about that?

2015-07-09 03:17:21 -0500 received badge  Editor (source)
2015-07-08 15:49:42 -0500 asked a question Best method to recognize objects?

Hi! I am trying to recognize traffic signs, I already detect them, but now I have to do something to know which one is each of them, I have been reading about the cascade classifier http://docs.opencv.org/doc/user_guide... but i would need 1 for each sign right? Furthermore i am having some issues with the merging of vec files. What do you recommend me to use for this object recognition? i am not sure if the cascade classifer is the best method.....

Thanks a lot fot your help!!

I also posted it here: http://stackoverflow.com/questions/31...

2015-07-08 14:54:42 -0500 received badge  Enthusiast
2015-07-05 02:14:36 -0500 commented question How to train an adaboost classifier?

ok, i know how to do it more of less know, but i have some doubts, I have the training images in .ppm extension, can I make the training with that format? I also have the information about the dataset like this: Filename;Width;Height;Roi.X1;Roi.Y1;Roi.X2;Roi.Y2;ClassId 00000_00000.ppm;56;57;6;5;51;52;1 00000_00001.ppm;55;57;5;6;50;52;1 00000_00002.ppm;56;56;5;6;51;51;1 00000_00003.ppm;58;58;6;6;53;52;1 00000_00004.ppm;59;59;6;6;54;53;1 00000_00005.ppm;57;59;5;6;52;54;1 00000_00006.ppm;57;59;5;5;52;54;1 00000_00007.ppm;57;59;5;5;52;53;1 May I use that or do I have to obtain new images an description files?

2015-07-03 15:23:51 -0500 asked a question How to train an adaboost classifier?

Hi, I intend to difference between different object by using the haar like base adaboost classifer, i know it only says if it is the object or it is not, so i would have to train one for each kind of object right? However I am not able to find a tutorial where I am tought how to do this, could you help me?

sorry for the incoveniece and thank you for your help!!!

2015-04-19 10:48:43 -0500 asked a question Is there other methods to detect circle apart of houghcircles

I am trying to detect circles road signs and I have some issues.

1.- The HoughCircles function detect circles in a gray image, however with the same parameters but the image binarized(the circle is still perfectly visible) it does not detect any circle, i do not why but with a binarized image it fails a lot, any idea why this issues with binary images?

2- to try to correct that i put the dp parameter in 2, and changed the threshold, in the binary image now i detect circles, but it also gives me a lot of false positives, i do not understand what it is the dp parameter or how to use it

3.- If you tell me that there is no way to make it work, i would like to know if there is any other way of detecting circles in an image.

crosspost on stackoverflow

thanks a lot for your help

2015-04-13 10:18:03 -0500 commented answer Detect not completed triangles

i have tried to do this and it detects more triangles(when some parts of the sides are missing it sometimes still fail) however now i get a lot of false positives in the images, in parts where there is nothing remotly similar to a triangle

2015-04-06 09:22:38 -0500 asked a question Detect not completed triangles

I am tryng to detect traffic signs by their colour and shape, i have the binary image from de colour segmentation and i try to detect triangles in it, by unless all the sides of the triangle are complete y doesnt work,

in this image for example it does not detect any triangle

the code i am using is this:

vector<Point> approx;
   findContours(copia,contours,CV_RETR_EXTERNAL,CV_CHAIN_APPROX_SIMPLE);
   vector<Rect_ <int> > boundRect( contours.size() );
   vector<Rect_ <int> > boundRect_( contours.size() );
   for(size_t i=0; i<contours.size();i++)
   {

       approxPolyDP(Mat(contours[i]), approx,arcLength(Mat(contours[i]), true)*0.02, true);

       if (approx.size() == 3 &&fabs(contourArea(Mat(approx))) > 300 && isContourConvex(Mat(approx)))
           drawContours(capture->image,contours,i,Scalar(0,255,0));

It is not a problem of the cotourArea, i have no problem detecting complete triangles. I do not know if i could do something to detect that kind of "not complete triangles" or i would have to do something to complete the triangle so that i could detect them. In either case i am a bit lost

Thank you for your help

Edit: Crosspost at Stackoverflow: http://stackoverflow.com/questions/29...

2015-02-21 03:45:04 -0500 asked a question How to play a .mov from a rtsp

Hi, i am trying to play from a wifi camera that streams on .mov file using rtsp, i get the image but it arrives distorsionated, sometimes pixelates sometimes green and sometimes without errors, while i am playing it i get the following messages in the bash:

    left block unavailable for requested intra mode at 0 29
[h264 @ 0x1e5af00] error while decoding MB 0 29, bytestream (-1)
[h264 @ 0x1e5b4c0] left block unavailable for requested intra mode at 0 29
[h264 @ 0x1e5b4c0] error while decoding MB 0 29, bytestream (-1)
[h264 @ 0x1e56900] left block unavailable for requested intra mode at 0 28
[h264 @ 0x1e56900] error while decoding MB 0 28, bytestream (-1)
[h264 @ 0x1e56900] left block unavailable for requested intra mode at 0 28
[h264 @ 0x1e56900] error while decoding MB 0 28, bytestream (-1)
[h264 @ 0x1ef56e0] left block unavailable for requested intra mode at 0 29
[h264 @ 0x1ef56e0] error while decoding MB 0 29, bytestream (-1)
[h264 @ 0x1e5af00] left block unavailable for requested intra mode at 0 19

to see the images the camera captures i am using this

 cv::VideoCapture capture("rtsp://192.168.1.254/sjcam.mov");
if (!capture.isOpened()) {
        //Error
    }

    cv::namedWindow("TEST", CV_WINDOW_AUTOSIZE);

    cv::Mat frame;

    for(int i =0; i<50000;i++) {
        if (!capture.read(frame)) {
            //Error
        }
        cv::imshow("TEST", frame);

        cv::waitKey(30);
    }

I do not know what else can i do, or where is the problem, i have tried to play .mov videos using opencv and i do it without problems, so i suppose i am doing something wrong with the rtsp thanks a lot